Genetic Algorithm Based Design Optimization of a Passive Anti-Roll Tank in a Sea Going Vessel

This paper presents the development of a numerical optimization tool for the design of a passive U-tube type anti-roll tank (ART) system to mitigate the roll motions of a vessel. A Genetic Algorithm (GA) based optimization scheme has been developed to minimize the objective function, which in the present investigation is chosen to be the area under the roll response transfer function curve. A practical nonlinear time-domain based body-exact strip theory technique is used to solve the fully coupled ART–ship motion problem. In the optimization process, the GA is linked to the fluid solver to approach an optimum ART design. The optimization variables are chosen to be the principal dimensions of the anti-roll tank, which have bearing on the roll reduction characteristics of the vessel. The study is implemented in the case of a coastal research vessel requiring passive stabilization via an ART. The results are validated using laboratory scale models to quantify the tank dynamics and effectiveness in isolation as well as when deployed in the vessel model and subjected to waves. The results demonstrate the validity and efficiency of the entire computational scheme. •An computational method has been developed to solve the coupled ship-ART problem.•The ART equations of motion have been derived using Lagrangian mechanics.•Scale model experiments have been conducted to validate computational model.•A Genetic Algorithm (GA) based (constrained) optimization routine has been developed.•The GA optimized ART form was found to reduce roll by approximately 12%.